Tuberculosis (TB) is the most frequent cause of death from a single infectious agent. Active disease follows about 5% of exposures, but most develop a 'latent' infection without symptoms that may reactivate later, particularly during immune deficiency. Live M. bovis bacillus Calmette-Guerin (BCG) is the only currently available vaccine but has had negligible impact on the global epidemic. Indeed, the rise of HIV infection has raised serious safety concerns, with disseminated infection reported in HIV-seropositives following BCG vaccination. Clearly there is an urgent need for safer, more effective vaccination for protective responses that, ideally, will prevent or contain latent TB infection. We have recently shown that consecutive immunization with DNA vaccines and attenuated fowlpoxvirus vectors encoding similar vaccine antigens generates high levels of antigen-specific, interferon-gamma-secreting CD4+ and CD8+ T cells that, importantly, exhibit markedly increased sensitivity (avidity) for the immunizing antigen. In terms of memory, antigen challenge led to rapid expansion of systemic and mucosal T cell effectors in vaccinated animals, reaching levels as high as 30% of total T cell numbers. Mucosal T cells may be particularly important in pulmonary TB, where little attention has been paid to local immune responses. Here, our primary aim is to test the hypothesis that T cell responses induced by prime-boost vaccination will control primary TB infection and the establishment or reactivation of latent infection. Our goal is to generate protective T cell responses against key antigens of M. tuberculosis (MTb) normally expressed (i) in acute infection and (ii) during progression to a nonreplicating persistent state. The latter are poorly studied but are of central importance to our proposal, since they represent novel and highly selective targets for vaccine strategies that may forestall or contain latent infection. Specifically, we will test the protective efficacy of systemic prime-boosting against (i) low-dose aerosol challenge with MTb and (ii) reactivation in murine models of latent TB disease. We will also study (iii) the protective capacity of prime-boost vaccines given mucosally and characterize systemic and mucosal (pulmonary) CD4+ and CD8+ T cell responses in these models. Co-delivery of IL-15 genes will be tested for their ability to enhance magnitude and memory of T cell immunity. The capacity of our approach to generate strong, sustained Th1-type CD4+ and CD8+ T cell responses is highly attractive in the context of TB and HIV infection, where maintenance of such responses against key proteins of MTb may be critical for protection.